The high mortality of ovarian cancer (OvCa) is caused by the wide dissemination of cancer cells within the abdominal cavity, which results in significant tumor burden. The first cells which metastatic OvCa cells encounter are the mesothelial cells which line the peritoneum and omentum of the abdominal cavity. We established an organotypic 3D culture of the peritoneum and omentum using primary human fibroblasts, mesothelial, and OvCa cells that allowed us to mimic the early steps of metastasis in a cell culture. Following the attachment of OvCa cells to the peritoneum, cancer cell derived MMP-2 cleaves fibronectin into smaller pieces which promotes ?51-integrin mediated adhesion and metastasis. c-Met supports the metastasis of OvCa cells through its physical with ?51-integrin, which allows the HGF-independent activation of c-Met/Src/Fak intracellular signaling. Additional mechanisms which promote OvCa dissemination involve the interaction of ?51-integrin with the urokinase receptor (uPAR) Most of the results obtained during the previous funding period point towards a prominent role for the ?51-integrin - fibronectin interaction in ovarian cancer metastasis. This is also supported by our clinical studies, which established that 90% of all OvCa metastases express fibronectin in the stroma. Because our preliminary data suggest that mesothelial cells are the source of fibronectin, the primary hypothesis underlying this application is that OvCa cells stimulate human mesothelial cells to secrete fibronectin, which increases OvCa invasion and metastasis. In Aim I we will elucidate the mechanisms with which the interaction of mesothelial cells with OvCa cells promotes invasion and metastasis through secretion of fibronectin. Experiments will include studying the role of fibronectin in vivo using a floxed fibronectin knock-out mouse which allows deletion of fibronectin in mesothelial cells, analysis of fibronectin promoter regulation in mesothelial cells, and a siRNA screen to understand how OvCa cells promote fibronectin secretion in the mesothelial cells. In Aim II we will perform a detailed study to identify which fibronectin fragments promote the association between ?51-integrin and c-Met. This will be followed by deletion of functional domains in ?5-integrin and c-Met to define the binding sites where the two proteins interact and the biological significance of this protein-protein interaction for the intra cellular signaling, invasion, and metastasis of OvCa cells. Based on the findings in Aims I and II, in Aim III, we will combine a c-Met inhibitor with a fibronectin peptide known to block the effect of fibronectin binding integrins and test them in pre-clinical treatment studies using an orthotopi xenograft model (bursal injection of primary cells) as well as in a genetic model of OvCa. Successful inhibition of fibronectin function could result in a novel and clinically relevant approach to the treatment of metastatic ovarian cancer.